How to Make a Plan: what to do when there is no tactic

A plan is not invented, it is read off the board. When there is no tactic in the position, the method has three steps: evaluate the position, find the imbalance, then choose a plan that uses your advantage against your opponent's weakness. That is how you get through a middlegame.

Evaluate before you look for moves

Before you calculate, take inventory. Five questions, always in the same order:

1. Material: who has more? The player ahead trades pieces; the player behind keeps things messy.
2. King safety: have both kings castled? A king in the centre cancels every slow plan.
3. Pawn structure: doubled, isolated, backward and passed pawns, squares no pawn defends. Pawns never move back, so this is the position's most permanent feature.
4. Piece activity: do not count pieces, look at what they do. A rook on an open file is worth far more than the same rook buried on a1.
5. Space: who owns each half of the board? More space means faster manoeuvring and fewer squares for the other side.

With practice this takes half a minute. And the answer is almost never "it is equal": one of the five items comes out lopsided, and that item becomes the plan.

The position comes from the Queen's Gambit Declined after 1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.cxd5 exd5, and the checklist says: material level, kings safe, development and space comparable. One item is left over. On the queenside White has two pawns (a2 and b2) against Black's three (a7, b7 and c6). That is the imbalance.

White plays b4, a4 and b5. Two pawns against three sounds like bad arithmetic, and that is the point: once the b-pawn trades itself on c6, Black is left with a backward pawn on c6 on a half-open file, or an isolated pawn on d5. A fixed target, white rooks on the c-file. This is the minority attack. Black, with nothing to attack over there, plays with pieces on the kingside (...Ne4, ...f5) or prepares the ...c5 break.

Nobody invented anything: both plans came straight out of the pawn structure.

The imbalance dictates the plan

The starting position is symmetrical and offers nobody a plan. Every trade and every pawn advance creates a difference, and the difference tells you what to do. The rule fits in one line: play where you are stronger and attack what your opponent cannot move.

The four standard plans

Almost every sound middlegame plan is one of the four ideas below. Once your evaluation has found an imbalance, one of them already is your plan.

• Attack where you have more space. On the wing where you are cramped, defend and trade pieces.
• Play towards the side your pawn chain points. With a locked centre, the chain shows which way you advance and the base of the enemy chain shows you the target.
• Open lines for your better pieces. Bishops want an open board, rooks want files, knights want a support square. Before a pawn break, ask who profits from the lines that open.
• Trade into a favourable endgame. A small permanent edge (a passed pawn, a healthier structure, a good bishop against a bad one) grows once the queens come off.

From here the game almost writes itself. Black piles up on d4: ...c5, ...Nc6, ...Qb6, often with ...Nge7 heading for f5. White plays on the other wing, where the e5 pawn gives space: Bd3, f4 and the advance f5 against e6. Whoever reaches the enemy base first wins the argument.

Trading pieces is a plan too

Plenty of players go through a whole middlegame without asking the cheapest question in chess: is the endgame that comes out of this good for me? If it is, every trade becomes progress.

White pushes the rook pawn and the black king has to cross the whole board to collect it. Meanwhile the white king walks in on the other side: 1.a5 Kd6 2.a6 Kc6 3.a7 Kb7 4.Ke5 Kxa7 5.Kf6, and f7, g7 and h7 fall one by one. An outside passed pawn is worth far more in the endgame than in the middlegame, and seeing that twenty moves earlier turns every exchange into part of the plan.

Always ask what your opponent wants

Half of all bad plans are good plans that ignored the reply. After every move your opponent makes, before you go back to your own idea, ask three questions:

• What does that move attack? The obvious question, and the one most often skipped in a fast game.
• What did it stop defending? Every piece that leaves a square abandons something. That is where tactics come from.
• If they moved twice in a row, what would they do? This question finds the real threats. If the answer is serious, your plan waits.

From plan to move

A good plan is short and concrete. "Reroute the knight from f3 to e5 via d2 and c4" is a plan: it has a beginning, a middle and an end. "Attack the king" is a wish. Most real plans have two or three steps, and you re-evaluate at the end of each one.

And when the evaluation turns up nothing? There is a waiting plan that is almost never wrong: improve your worst piece. Find the piece doing the least work and give it a better square.